Scores by kind of claim
Each kind of claim is scored on its own, out of 100: the share that held up, with the range it likely falls in. Fewer resolved claims give a wider range.
-
Factual claims
Statements about the present or the past, checked against the record: did they hold up?
78 72 83 -
Predictions
Statements about events outside the person's control, checked once they fell due: did they come true?
64 43 83 -
Promises
Commitments to the person's or organisation's own future action, with something to deliver: was it delivered?
Insufficient record: 0 promises resolved, 5 needed for a score.
All types combined: 77 (likely 71 to 82, high confidence), from 84 claims: 79 factual claims, 5 predictions. It mixes kinds of claim in different proportions for different people, so compare people kind by kind.
Yann LeCun's claims
3 claims
| Verdict | Kind | Said | Claim | Topic |
|---|---|---|---|---|
| Mixed | Factual | 8 Mar 2024 | For the last 12 to 15 years, people have claimed AGI is just around the corner and have been systematically wrong. Factual · 8 Mar 2024 | AI timelines and forecasting |
| Mixed | Factual | 27 May 2015 | Deep learning methods have dramatically improved the state of the art in many other domains such as drug discovery and genomics. Factual · 27 May 2015 | AI capabilities |
| Mixed | Factual | 2010 | ConvNets have been successfully deployed in many commercial applications from OCR to video surveillance, but require large amounts of labeled training samples. Factual · 2010 | AI capabilities |
“many other domains such as drug discovery and genomics”
LeCun, Bengio and Hinton, article abstract 27 May 2015 Deep learning (Nature) (Nature)
Deep learning methods have dramatically improved the state of the art in many other domains such as drug discovery and genomics.
Mixed Partly true and partly false in roughly equal measure.
What would count as true or false
True if published results as of May 2015 show deep learning methods markedly improving the state of the art in drug discovery (for example the Merck Kaggle challenge, Tox21) and in genomics (for example predicting splicing or the effects of mutations), as well as in other domains. Mostly true if clear improvements are shown in both but 'dramatically' overstates the gains in one of them. Mixed if clear state-of-the-art gains are documented for only one of drug discovery and genomics. Mostly false if the evidence by then amounts to only early or modest results in both, stated as dramatic. False if deep learning had not improved the state of the art in either field by then.
Evidence
- Toxicity Prediction using Deep Learning arXiv (ar5iv rendering); authors at Institute of Bioinformatics, Johannes Kepler University Linz
“Though Deep Learning is new to the field and was never applied to toxicity prediction before, it clearly outperformed all other participating methods.”
- Deep Neural Nets as a Method for Quantitative Structure−Activity Relationships Journal of Chemical Information and Modeling (Ma, Svetnik, Sheridan, Dahl, Liaw; Merck Research Laboratories and University of Toronto)
“Here we show that DNNs can routinely make better prospective predictions than RF on a set of large diverse QSAR data sets that are taken from Merck’s drug discovery effort.”
- Deep Neural Nets as a Method for Quantitative Structure−Activity Relationships Journal of Chemical Information and Modeling (Ma, Svetnik, Sheridan, Dahl, Liaw; Merck Research Laboratories and University of Toronto)
“The winning entry (submitted by one of the authors, George Dahl) improved the mean R2 averaged over the 15 data sets from 0.42 (for RF) to 0.49.”
- Predicting the sequence specificities of DNA- and RNA-binding proteins by deep learning - Nature Biotechnology Nature Biotechnology · 27 July 2015
“Using a diverse array of experimental data and evaluation metrics, we find that deep learning outperforms other state-of-the-art methods, even when training on in vitro data and testing on in vivo data.”
Reasoning
For drug discovery, the evidence shows clear gains. The toxicity prediction paper (E1, arXiv 1503, which is the March 2015 numbering) says deep learning "clearly outperformed all other participating methods" in toxicity prediction. The Merck QSAR paper (E2, E3) reports that deep neural nets routinely beat random forests, and that the winning Kaggle entry raised mean R² from 0.42 to 0.49, a clear but moderate gain. The only genomics evidence (E4) was published on 2015-07-27, after the 2015-05-27 judge date, so the evidence does not document genomics gains as of the claim date. Since state-of-the-art gains are documented for only one of the two named fields, this falls under the criteria's 'mixed' tier.
Cite this claim
Yann LeCun, “Deep learning methods have dramatically improved the state of the art in many other domains such as drug discovery and genomics.” (said 27 May 2015). Verdict: mixed. Reliability API, claim c_92a884f7acba, method 0.1.0+c876030c. https://epistacklabs.com/claims/c_92a884f7acba
Claims without a verdict
- Unresolvable 110
- Not yet due 3
- Not researched 309
- Not checkable 669
Is a verdict here wrong, or do you speak for Yann LeCun? Reply on the claim's own page, with the form at its foot (how replies work). Verdicts are made automatically; how verdicts and scores are made.